Creep Test and Accuracy Test results

Joe_Tanner

Hello, 

We’re with Octopus and since Summer last year our energy usage more than doubled without any changes in our habits or new appliances being bought. At first I thought that it was just the energy prices going crazy but upon analysing our submitted readings a few months ago I noticed there was a month when there was a massive jump in our usage to more than double than usual and since then it remained like that every month. 

I contacted Octopus and they advised to perform a creep test on our non-smart digital meter. I did the creep test on two separate occasions, turned off the fuse box main switch and also the big switch on the cable that goes from the meter into the fuse box for thirty minutes and both times the red light kept blinking and the first time it measured 0.4 units and the second time 0.5.

Octopus decided to send an engineer to perform an Onsite Accuracy Test and after more than four months waiting and an Ombudsman complaint he finally came over and did the test. He told us that Octopus didn’t want him to discuss the results with us, which I found a bit odd, but after a week received an email saying that the meter had been found faulty. I called customer service and they told me it was running 5.49% fast and that our bills would be adjusted based on that value. That sounds like a very small inaccuracy and adjustment compared to how much our usage has gone up.

What I’d like to understand is if the results of the creep test indicate a large or a small anomaly and if it’s possible to establish a relation between that and the result from the accuracy test? O.4 units in thirty minutes sounds like a lot to me if you add it up for a whole day and 5.49% doesn't sound like much compared to that, but I might be wrong if there’s another way to look at it. I really don’t know how to interpret the results and would like to understand if it’s worth it to dispute them as we’re now waiting for a booking for a smart meter to be installed and I’m a bit worried that once the old meter’s been removed there won’t be much we can do.

Appreciate any help.

Thanks.

Krakkkers

Are you talking about a doubling of the cost or a doubling of the usage in kwh?

BarelySentientAI

Joe_Tanner said:

What I’d like to understand is if the results of the creep test indicate a large or a small anomaly and if it’s possible to establish a relation between that and the result from the accuracy test?

Not really any good relationship. 

The creep test is designed to show whether the meter has an 'offset' type error - which adds - so in your example it could be understood to have a offset of 0.8 to 1.0 kW at zero load.  The problem with that is you don't know if the offset changes when there is current going through the meter.  It proves that the meter is defective, because there should be no offset, but doesn't show how wrong the reading is.  It's unlikely that you were very often using completely no electricity.

An accuracy test shows the percentage error on the reading - which multiplies - so if you put 1 unit through the meter, how many units does the reading change by.  Going up by 1.1 would be a +10% error, going up by 0.9 would be a -10% error.  Again, a single test would only show the error for one particular load, and meters don't respond exactly the same to all different loads (so might be slow when there is not much switched on and fast when lots of things are turned on, or vice versa, or something else entirely).  There is allowed to be a few % error at all different points for a meter to still be 'accurate'.

The next step you could take would be to ask for the meter to be replaced and the old one tested for accuracy.  That would involve multiple tests in an independent lab to see how accurate the meter is overall.  I don't know if they would come up with a different number from Octopus though.  https://www.gov.uk/guidance/electricity-meter-accuracy-disputes   

Reed_Richards

If your meter is going up 0.4 kWh in 30 minutes that's 0.8 units per hour or 19.2 units per day.  That would add maybe £4 per day to your bill!

It seems to me that if your meter was creeping up when no electricity was being drawn then it becomes much more difficult to assess if it was also running fact.  And if it failed the creep test then Octopus need to allow for that in their calculations.

If I were you I would try to measure the rate of creep more accurately, perhaps over a longer time period.  Take photos to show what you have done.  Then insist that Octopus adjust their bills for creep as well as running fast.  Alternatively, if nothing in your house has changed, insist that Octopus base their bills on the new readings from your smart meter assuming that last year's readings, had they been recorded accurately, would have been the same.       

BarelySentientAI

Reed_Richards said:

If your meter is going up 0.4 kWh in 30 minutes that's 0.8 units per hour or 19.2 units per day.  That would add maybe £4 per day to your bill!

It seems to me that if your meter was creeping up when no electricity was being drawn then it becomes much more difficult to assess if it was also running fact.  And if it failed the creep test then Octopus need to allow for that in their calculations.

If I were you I would try to measure the rate of creep more accurately, perhaps over a longer time period.  Take photos to show what you have done.  Then insist that Octopus adjust their bills for creep as well as running fast.  Alternatively, if nothing in your house has changed, insist that Octopus base their bills on the new readings from your smart meter assuming that last year's readings, had they been recorded accurately, would have been the same.       

That's not how it works.  And nor should it be.

One meter that I tested in the past had a creep test result of 300W, but it turned out that it was acting as a minimum reading, so at no load it recorded 300W, at 100W load it recorded 300W, at 200W load it recorded 300W, and from 300W and upwards it was accurate.

Given that most properties would be using at least some electricity all of the time, applying creep as well as accuracy test results would have been a massive overestimate of the error in that case.

mmmmikey

Just a thought.....

If you have a record of your usage over the last few years and it's fairly consistent, then another way to go might be to ask Octopus to base your bills on an estimate, using those readings.

It would be an easier argument to make as it avoids getting bogged down in the technicalities.

MP1995

I wonder if the meter reading when switching to octopus was also correct or processed incorrectly.

Reed_Richards

BarelySentientAI said:

Reed_Richards said:

If your meter is going up 0.4 kWh in 30 minutes that's 0.8 units per hour or 19.2 units per day.  That would add maybe £4 per day to your bill!

It seems to me that if your meter was creeping up when no electricity was being drawn then it becomes much more difficult to assess if it was also running fact.  And if it failed the creep test then Octopus need to allow for that in their calculations.

If I were you I would try to measure the rate of creep more accurately, perhaps over a longer time period.  Take photos to show what you have done.  Then insist that Octopus adjust their bills for creep as well as running fast.  Alternatively, if nothing in your house has changed, insist that Octopus base their bills on the new readings from your smart meter assuming that last year's readings, had they been recorded accurately, would have been the same.       

That's not how it works.  And nor should it be.

One meter that I tested in the past had a creep test result of 300W, but it turned out that it was acting as a minimum reading, so at no load it recorded 300W, at 100W load it recorded 300W, at 200W load it recorded 300W, and from 300W and upwards it was accurate.

Given that most properties would be using at least some electricity all of the time, applying creep as well as accuracy test results would have been a massive overestimate of the error in that case.

So @BarelySentientAI under those circumstances how on earth do you estimate how much the meter was over-reading?  Surely the only thing you can do is look at pre and post fault readings and assume that during would have been much the same.  That was one suggestion I offered and @mmmmikey said the same.    

BarelySentientAI

Reed_Richards said:

BarelySentientAI said:

Reed_Richards said:

If your meter is going up 0.4 kWh in 30 minutes that's 0.8 units per hour or 19.2 units per day.  That would add maybe £4 per day to your bill!

It seems to me that if your meter was creeping up when no electricity was being drawn then it becomes much more difficult to assess if it was also running fact.  And if it failed the creep test then Octopus need to allow for that in their calculations.

If I were you I would try to measure the rate of creep more accurately, perhaps over a longer time period.  Take photos to show what you have done.  Then insist that Octopus adjust their bills for creep as well as running fast.  Alternatively, if nothing in your house has changed, insist that Octopus base their bills on the new readings from your smart meter assuming that last year's readings, had they been recorded accurately, would have been the same.       

That's not how it works.  And nor should it be.

One meter that I tested in the past had a creep test result of 300W, but it turned out that it was acting as a minimum reading, so at no load it recorded 300W, at 100W load it recorded 300W, at 200W load it recorded 300W, and from 300W and upwards it was accurate.

Given that most properties would be using at least some electricity all of the time, applying creep as well as accuracy test results would have been a massive overestimate of the error in that case.

So @BarelySentientAI under those circumstances how on earth do you estimate how much the meter was over-reading?  Surely the only thing you can do is look at pre and post fault readings and assume that during would have been much the same.  That was one suggestion I offered and @mmmmikey said the same.    

Use a 'typical' demand distribution (or in the case of a smart meter an actual one from a period prior to any fault) and calculate on that basis.  It relies on assumptions and is of course not perfectly accurate, but is theoretically at least including the right weather trends and general magnitude of use.

Or try and argue that future use is the same as past use.  Assumptions and inaccuracies there too, neither are perfect methods.

Chris_b2z

What is the make and model? Also, does it have a label showing year of manufacture?

Joe_Tanner

Krakkkers said:

Are you talking about a doubling of the cost or a doubling of the usage in kwh?

Doubling of usage. more than doubling actually. We're in a two bed flat and in 2021 we've used 2455 units in total, in 2022 we've used 2284 units, and so far in 2024 we've already used 4185 units. Problem seems to have started in September 2023, when usage jumped from 285 units in August to 428 in September. And in October we've been away for two weeks and it still measured 449 units during that month.